Re: [boost] boost-1.28.0 failure with gcc-3.1

From: Neal D. Becker (nbecker_at_[hidden])
Date: 2002-05-17 08:21:52

>>>>> "Matthias" == Matthias Troyer <troyer_at_[hidden]> writes:

    Matthias> On Friday, May 17, 2002, at 02:53 PM, David Abrahams wrote:

    >> Are there  one or more GCC-3.1 bugs uncovered here?
    >> Has anyone submitted a report to gnats?
    >> http://gcc.gnu.org/cgi-bin/gnatsweb.pl
    >> 
    >> ----- Original Message -----
    >> From: "Neal D. Becker" <nbecker_at_[hidden]>
    >> To: <boost_at_[hidden]>
    >> Sent: Friday, May 17, 2002 7:36 AM
    >> Subject: Re: [boost] boost-1.28.0 failure with gcc-3.1
    >> 
    >> 
    >>> Thanks!  That also fixed it here.  Strange that my completely bogus
    >>> patch also silenced gcc's complaints.

    Matthias> I don't think so. According to the standard (Sec. 24) ostream_iterator
    Matthias> is defined

    Matthias> in the <iterator> header. This is probably implicitly included by some
    Matthias> other

    Matthias> header on other platforms, but not on gcc-3.1 . It was thus a bug in
    Matthias> boost::random.

But isn't it a bug that gcc accepted my bogus code?